Performance of 3D Morphological Methods in the Machine Learning Assisted Classification of Closely Related Fossil Bivalve Species of the Genus Dreissena

Klinkenbuß, Denise; Metz, Olivia; Reichert, Jessica; Hauffe, Torsten; Neubauer, Thomas; Wesselingh, Frank P.; Wilke, Thomas

doi:10.4002/040.063.0109

Cited by 5 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the future, it may be possible to use this feature to find similar classification boundaries relying on models to perceive more detailed information about fossils ( e.g. , ornamental features and 3D-morphology), which in turn could allow for quantitative differentiation of gradual features ( Klinkenbußet al, 2020 ; Edie, Collins & Jablonski, 2023 ). That could not only provide new possible perspectives for exploring fossil classification and biomorphological evolution, but also try to explore whether there are important features that have been overlooked by experts.…”

Section: Discussionmentioning

confidence: 99%

“…CNNs can complement existing methods for morphological studies such as morphological matrix ( Dai, Korn & Song, 2021 ), landmark ( Bazzi et al, 2018 ), fractal dimensions ( Wiese et al, 2022 ), ornamentation index ( Miao et al, 2022 ), conch properties ( De Baets, 2021 ), and 3D morphological methods ( Klinkenbußet al, 2020 ) and provide new perspectives for studying the morphological evolution of fossils in the future. Geometric morphometry requires the extraction of fossil features by labelling manually and performing descending operations ( e.g.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Automatic identification and morphological comparison of bivalve and brachiopod fossils based on deep learning

Sun,

Liu,

Huang

et al. 2023

PeerJ

View full text Add to dashboard Cite

Fossil identification is an essential and fundamental task for conducting palaeontological research. Because the manual identification of fossils requires extensive experience and is time-consuming, automatic identification methods are proposed. However, these studies are limited to a few or dozens of species, which is hardly adequate for the needs of research. This study enabled the automatic identification of hundreds of species based on a newly established fossil dataset. An available “bivalve and brachiopod fossil image dataset” (BBFID, containing >16,000 “image-label” data pairs, taxonomic determination completed) was created. The bivalves and brachiopods contained in BBFID are closely related in morphology, ecology and evolution that have long attracted the interest of researchers. We achieved >80% identification accuracy at 22 genera and ∼64% accuracy at 343 species using EfficientNetV2s architecture. The intermediate output of the model was extracted and downscaled to obtain the morphological feature space of fossils using t-distributed stochastic neighbor embedding (t-SNE). We found a distinctive boundary between the morphological feature points of bivalves and brachiopods in fossil morphological feature distribution maps. This study provides a possible method for studying the morphological evolution of fossil clades using computer vision in the future.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Automatic identification and morphological comparison of bivalve and brachiopod fossils based on deep learning

Sun,

Liu,

Huang

et al. 2023

PeerJ

View full text Add to dashboard Cite

show abstract

“…Based on Mandelbrot ( 1982 ) and his concept of fractal geometry, another more secret “revolution in morphometrics” may pick up speed despite the criticism “that a fractal cow is often not much better than a spherical cow ” (Buldyrev, 2012 ). Quite a few studies across (paleo‐) biological disciplines have demonstrated the potential of fractals for morphometrics (Aiello et al, 2007 ; Bruno et al, 2008 ; Isaeva et al, 2006 ; Klinkenbuß et al, 2020 ; Lutz & Boyajian, 1995 ). Kaczor et al ( 2012 ) suggested fractal dimensions as an indicator of roughness in protein structures.…”

Section: Introductionmentioning

confidence: 99%

Can fractal dimensions objectivize gastropod shell morphometrics? A case study from Lake Lugu (SW China)

Wiese

Harrington

Hartmann

et al. 2022

Ecology and Evolution

View full text Add to dashboard Cite

Morphometrics are fundamental for the analysis of size and shape in fossils, particularly because soft parts or DNA are rarely preserved and hard parts such as shells are commonly the only source of information. Geometric morphometrics, that is, landmark analysis, is well established for the description of shape but it exhibits a couple of shortcomings resulting from subjective choices during landmarking (number and position of landmarks) and from difficulties in resolving shape at the level of micro‐sculpture. With the aid of high‐resolution 3D scanning technology and analyses of fractal dimensions, we test whether such shortcomings of linear and landmark morphometrics can be overcome. As a model group, we selected a clade of modern viviparid gastropods from Lake Lugu, with shells that show a high degree of sculptural variation. Linear and landmark analyses were applied to the same shells in order to establish the fractal dimensions. The genetic diversity of the gastropod clade was assessed. The genetic results suggest that the gastropod clade represents a single species. The results of all morphometric methods applied are in line with the genetic results, which is that no specific morphotype could be delimited. Apart from this overall agreement, landmark and fractal dimension analyses do not correspond to each other but represent data sets with different information. Generally, the fractal dimension values quantify the roughness of the shell surface, the resolution of the 3D scans determining the level. In our approach, we captured the micro‐sculpture but not the first‐order sculptural elements, which explains that fractal dimension and landmark data are not in phase. We can show that analyzing fractal dimensions of gastropod shells opens a window to more detailed information that can be considered in evolutionary and ecological contexts. We propose that using low‐resolution 3D scans may successfully substitute landmark analyses because it overcomes the subjective landmarking. Analyses of 3D scans with higher resolution than used in this study will provide surface roughness information at the mineralogical level. We suggest that fractal dimension analyses of a combination of differently resolved 3D models will significantly improve the quality of shell morphometrics.

show abstract

“…While in the past this was done by hand or extracting data from two-dimensional photos and illustrations, high-throughput techniques such as magnetic resonance imaging (MRI), computed tomography (CT) scanning, structured light scanning, and photogrammetry have made it possible to capture morphology in digital and 3D data sets (e.g., Bythell et al, 2001;Faulwetter et al, 2013;Sigl et al, 2013;Reichert et al, 2016). Alternative descriptors of 3D shape and complexity, such as fractal dimension and alpha shapes, have emerged as potential approaches for quantifying morphology in complex-shaped organisms and structures (Martin-Garin et al, 2007;Reichert et al, 2016;Gardiner et al, 2018;Klinkenbuß et al, 2020;Orbach et al, 2021). Yet, previous frameworks to extract meaningful characters in the absence of identifiable landmarks and characterize phena in complex modular organisms have either gauged only a few variables from 3D-morphological data (e.g., Gutierrez-Heredia et al, 2016;Reichert et al, 2017) or been restricted to twodimensional analyses (e.g., Reeb et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Quantitative three-dimensional morphological analysis supports species discrimination in complex-shaped and taxonomically challenging corals

et al. 2022

Self Cite

View full text Add to dashboard Cite

Morphological characters play an important role in species descriptions and are essential for a better understanding of the function, evolution and plasticity of an organism’s shape. However, in complex-shaped organisms lacking characteristic features that can be used as landmarks, quantifying morphological traits, assessing their intra- and interspecific variation, and subsequently delineating phenotypically distinct groups continue to be problematic. For such organisms, three-dimensional morphological analysis might be a promising approach to differentiate morphogroups and potentially aid the delineation of species boundaries, though identifying informative features remains a challenge. Here, we assessed the potential of 3D-based quantitative morphology to delineate a priori and/or to discriminate a posteriori morphogroups of complex-shaped and taxonomically challenging organisms, such as corals from the morphologically diverse genus Acropora. Using three closely related coral taxa previously delimited using other lines of evidence, we extracted a set of variables derived from triangulated polygon meshes and medial axis skeletons of the 3D models. From the resulting data set, univariate and multivariate analyses of 3D-based variables quantifying overall shape including curvature, branching, and complexity were conducted. Finally, informative feature selection was performed to assess the discriminative power of the selected variables. Results revealed significant interspecific differences in the means of a set of 3D-based variables, highlighting potentially informative characters that provide sufficient resolution to discriminate morphogroups congruent with independent species identification based on other lines of evidence. A combination of representative features, remarkably represented by curvature, yielded measures that assisted in differentiating closely related species despite the overall morphospaces overlap. This study shows that a well-justified combination of 3D-based variables can aid species discrimination in complex-shaped organisms such as corals and that feature screening and selection is useful for achieving sufficient resolution to validate species boundaries. Yet, the significant discriminative power displayed by curvature-related variables and their potential link to functional significance need to be explored further. Integrating informative morphological features with other independent lines of evidence appears therefore a promising way to advance not only taxonomy but also our understanding of morphological variation in complex-shaped organisms.

show abstract

Performance of 3D Morphological Methods in the Machine Learning Assisted Classification of Closely Related Fossil Bivalve Species of the Genus Dreissena

Cited by 5 publications

References 25 publications

Automatic identification and morphological comparison of bivalve and brachiopod fossils based on deep learning

Automatic identification and morphological comparison of bivalve and brachiopod fossils based on deep learning

Can fractal dimensions objectivize gastropod shell morphometrics? A case study from Lake Lugu (SW China)

Quantitative three-dimensional morphological analysis supports species discrimination in complex-shaped and taxonomically challenging corals

Contact Info

Product

Resources

About